AI can help increase profit for utilities - with no impact to customers

It is a common misconception that utilities make money by selling more electricity - after all, most utility bills are based on kWh used. In reality, utilities are authorized annual “revenue requirements” by regulators, which for simplicity’s sake, is based on their rate base (capital additions) multiplied by an authorized rate of return, plus expenses (in Operations & Maintenance, taxes, deprecation, etc.). The authorized rate of return is determined by their state or federal regulators and can be thought of as the amount that a utility makes so that they can attract future investors.

Each year, utilities collect pre-determined revenue requirements from all customers. Since the revenue requirements are set, savings in O&M generally enable a utility to do more capital work, thereby allowing them to increase their net profits (associated with the rate base and rate of return). If the regulatory conditions are right, this can all happen while keeping their total revenue requirements the same, meaning no impact to customer bills.

AI can help utility companies reduce O&M costs by reducing the cost of data clean up projects through component detection and asset tag extraction, speeding up storm response and reducing storm costs, and cutting down on the cost of utility pole inspections through asset defect detection.

Reduce the expense of building analytics products

When utilities implement asset management using predictive analytics, they often require a significant amount of data. This means that companies will spend a good amount of time gathering, cleaning, and preparing data for those analytics products. According to a McKinsey report, industry leaders spend no more than 20 to 30 percent of the development time on this activity, but most utilities are spending 60 to 80 percent of the time, or up to 4X more time. AI can help reduce the development time required to build analytics products by geolocating assets, automatically detecting components, extract asset tags - leading to significant savings when implementing analytics or other data cleanup projects.

Reduce storm expenses

After a storm, utilities do damage assessments - inspecting and analyzing the impact of the storm before they begin to schedule the necessary capital and O&M work. Damage assessment can take anywhere from hours to days, depending on the extent of the damage and how widespread the damaged area is. By helping utilities speed up the process of damage assessments, AI can help utilities reduce the overtime needed from crews and assessors. Through analyzing images and video from aerial surveys, as well as ground surveys, AI can help to rapidly identify issues for back-off employees so that they can more quickly begin scheduling the necessary remediation work.

Reduce inspections expense

Combined, utilities spend billions every year inspecting their assets across their respective territories. And even as utilities begin to digitize and automate their asset inspection workflows, they often still perform manual, costly and time-consuming reviews of inspection imagery and video. Half of their overall inspection cost - or more - ends up being consumed by these manual reviews. AI can help quickly filter and sort through the photos and videos so that these manual reviewers can prioritize their review on images and video with issues - rather than waste valuable time scanning inspection data that do not have issues. This can help utilities reduce their manual review costs by up to 70%.

Summary

Adopting AI can result in a win for the utility as well as a win for the customer. AI can help utilities reduce O&M expenses, which impacts customers in the year in which they occur. By cutting O&M and increasing capital that generates value to the customer, utilities will be able to earn more of a return over time. Our next article will discuss how AI can help reduce inspections expenses in detail.

‍